| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NEUROSCIENCE |
1 Department of Integrative Physiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka 812-8582, Japan
2 Division of Histology and Neuroanatomy, Department of Anatomy and Physiology, Saga Medical School, Nabeshima 5-1-1, Saga 849-8501, Japan
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
- and C-fibres. Despite numerous anatomical and physiological studies, correlation between morphology and functional connectivity, particularly in terms of inhibitory inputs, remains elusive. To compare excitatory and inhibitory synaptic inputs on individual SG neurones with morphology, we performed whole-cell recordings with Neurobiotin-filled-pipettes in horizontal slices from adult rat spinal cord with attached dorsal roots. Based on dendritic arborization patterns, four major cell types were confirmed: islet, central, radial and vertical cells. Dorsal root stimulation revealed that each class was associated with characteristic synaptic inputs. Islet and central cells had monosynaptic excitatory inputs exclusively from C-afferents. Islet cells received primary-afferent-evoked inhibitory inputs only from A-fibres, while those of central cells were mediated by both A- and C-fibres. In contrast, radial and vertical cells had monosynaptic excitatory inputs from both A- and C-fibres and inhibitory inputs mediated by both fibre types. We further characterized the neurochemical nature of these inhibitory synaptic inputs. The majority of islet, central and vertical cells exhibited GABAergic inhibitory inputs, while almost all radial cells also possessed glycinergic inputs. The present study demonstrates that SG neurones have distinct patterns of excitatory and inhibitory inputs that are related to their morphology. The neurotransmitters responsible for inhibitory inputs to individual SG neurones are also characteristic for different morphological classes. These results make it possible to identify primary afferent circuits associated with particular types of SG neurone.
(Received 19 December 2006;
accepted after revision 5 March 2007;
first published online 8 March 2007)
Corresponding author T. Yasaka: Spinal Cord Group, West Medical Building, University of Glasgow, University Avenue, Glasgow G12 8QQ, UK. Email: t.yasaka{at}bio.gla.ac.uk
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Among the different laminae in the dorsal horn, the substantia gelatinosa (SG; lamina II of Rexed, 1952) is believed to play a critical role in nociceptive transmission, since finely myelinated (A) and unmyelinated (C) primary afferent fibres that convey predominantly nociceptive information terminate within it (Rexed, 1952; Kumazawa & Perl, 1978; Light & Perl, 1979; Yoshimura & Jessell, 1989a). The presence of numerous GABAergic and glycinergic interneurones within the SG, together with the inputs to it from other modulatory systems, including monoaminergic and opioid peptide-containing axons, also indicates that it has an essential role in controlling pain transmission (Todd & McKenzie, 1989; Todd & Sullivan, 1990; Hantman et al. 2004, Hantman & Perl, 2005; Heinke et al. 2004; Zeilhofer et al. 2005). The organization of the SG is very complex due to the morphological, neurochemical and electrophysiological heterogeneity of its constituent neurones (Ramon y Cajal, 1909; Pearson, 1952; Gobel, 1975, 1978; Beal et al. 1989; Yoshimura & Jessell, 1989b; Grudt & Perl, 2002; Todd & Koerber, 2005). There is a need for studies to investigate the relationship between these different properties for individual neurones in order to determine their role in regulating the flow of sensory information (particularly that from nociceptive afferents) through the dorsal horn.
In a combined electrophysiological and morphological study, Grudt & Perl (2002) categorized them into the four major cell groups in lamina II: islet, central, radial and vertical cells. Both islet and central cells had dendrites that were elongated in the rostrocaudal direction with limited mediolateral and dorsoventral spread, but the dendritic trees of islet cells were substantially longer than those of the central cells. The dendrites of radial cells extended in all directions, while those of the vertical cells had a significant dorsoventral spread and were predominantly ventral to the soma.
Little is known about the nature of the inhibitory inputs to each type of SG neurone, except that certain central cells receive input from GABAergic islet cells (Grudt & Perl, 2002; Lu & Perl, 2003). Thus, it is not known whether different morphological populations of SG neurone have characteristic types of inhibitory input. In this study, we examined the morphological features of SG neurones and categorized them into four major cell groups. We identified their characteristic excitatory and inhibitory responses to primary afferent stimulation, in order to provide further information about the neuronal circuits in which they were involved.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Preparation of the slice
The methods used for obtaining spinal cord slices from adult rats were similar to those previously described (Baba et al. 1994; Yang et al. 2001; Kato et al. 2004). Briefly, male Sprague-Dawley rats (7–8 weeks old) were deeply anaesthetized with urethane (1.2–1.5 g kg–1, I.P.). After thoracolumbar laminectomy at the level of T11 to L5, a 2.0–4.0 cm length of the spinal cord, with ventral and dorsal roots attached, was excised and placed in a preoxygenated cold (1–3°C) Krebs solution. The rats were killed by exsanguination. All of the ventral and dorsal roots, with the exception of the dorsal root(s) for stimulation (length, 1–2 cm) on the right side, were cut and the pia-arachnoid membrane was removed except around the preserved dorsal root. The spinal cord was placed in a shallow groove formed in an agar block. A horizontal slice (thickness, 500–650 µm) with attached dorsal root was cut with a Vibratome while the spinal cord was immersed in cold Krebs solution. The slice was mounted on a nylon mesh in the recording chamber, and perfused continuously at a flow rate of 15–20 ml min–1 with Krebs solution equilibrated with 95% O2 and 5% CO2 at 36 ± 1°C. The Krebs solution contained (mM): NaCl 117, KCl 3.6, CaCl2 2.5, MgCl2 1.2, NaH2PO4 1.2, NaHCO3 25, and glucose 11.
Recording and stimulation
Blind whole-cell voltage-clamp recordings were made from SG neurones, as previously described (Yoshimura & Nishi, 1993; Yang et al. 2001; Kato et al. 2004). The patch pipettes were filled with a solution containing (mM) Cs2SO4 110, TEA-Cl 5, CaCl2 0.5, MgCl2 2, EGTA 5, Hepes 5 and Mg-ATP 5, and cells were recorded in voltage-clamp mode. The tip resistance of the patch pipette was 8–12 M
. Signals were acquired with a patch-clamp amplifier (Axopatch 200B, Molecular Devices, Union City, CA, USA). Data were digitized with an analog-to-digital converter (Digidata 1321 A, Molecular Devices), stored on a personal computer using a data acquisition program (Clampex version 8.0, Molecular Devices), and analysed with Clampfit software (version 4.1, Molecular Devices). In voltage-clamp mode, cells were initially held at –70 mV (at which glycine- and GABA-mediated inhibitory postsynaptic currents, IPSCs, are negligible), and then at 0 mV (at which glutamate-mediated excitatory postsynaptic currents, EPSCs, are negligible) (Yoshimura & Nishi, 1993). Isolation of excitatory and inhibitory synaptic currents was further confirmed by applying appropriate antagonists. Stimuli (duration, 100 µs) to elicit EPSCs and IPSCs were applied to the dorsal root at a frequency of 0.2 Hz, via a suction electrode with intensities 1.2–1.5 times the threshold required to elicit a response in the most excitable A- or C-afferent fibres. The A- or C-afferent-mediated responses evoked by the dorsal root stimulation were distinguished on the basis of the conduction velocity (CV) of afferent fibres (C, < 0.8 m s–1; A, 2–11 m s–1) and stimulus threshold (C, > 200 µA; A, 40–200 µA), as previously described (Yoshimura & Jessell, 1989a; Nakatsuka et al. 1999; Ito et al. 2000; Kato et al. 2004). The precise CV was calculated by measuring the difference in latencies of responses evoked by two focal monopolar electrodes separately positioned on the root (Park et al. 1999; Kato et al. 2004). The A- and C-afferent-mediated EPSCs were considered as monosynaptic in nature when the latency remained constant during stimulation at 20 Hz for A-fibres and when failures did not occur irrespective of the changes of latency during repetitive stimulation at 2 Hz for C-afferents, respectively (Nakatsuka et al. 1999, 2000; Ataka et al. 2000; Ito et al. 2000). These criteria were based on the results obtained from dorsal root ganglion experiments (Ataka et al. 2000).
In this study, horizontal slices were used to investigate A- and/or C-afferent-mediated inputs to neurones in lamina II. It has been reported that in normal adult rats, A
-afferent-mediated responses can be elicited in lamina II neurones by dorsal root stimulation in transverse slices, but not in horizontal slices (Baba et al. 1999; Nakatsuka et al. 1999; Kato et al. 2004). This is presumably because many A-afferents pass ventrally through the dorsal horn, before turning dorsally to terminate in lamina III and the inner half of lamina II (lamina IIi), and are thus transected during preparation of the slice. In contrast, it has been shown that C-afferent-mediated IPSCs evoked by dorsal root stimulation are detected in horizontal slices but not in transverse slices (Kato et al. 2004). An additional reason for using horizontal slices was that since most lamina II neurones have dendritic trees with significant rostrocaudal spread, their morphology can be seen more clearly in horizontal slices than in transverse slices.
Identification of SG neurones
The superficial part of the dorsal horn, lateral to the dorsal column was distinguishable as a relatively translucent band below Lissauer's tract in the horizontal slice when observed under a binocular microscope (Baba et al. 1994; Kato et al. 2004). SG neurones were identified by the depth from the dorsal surface, and from their morphological features. Neurones were recorded at a depth of 30–120 µm from dorsal surface of the spinal cord, and in most cases their identity was further confirmed by labelling with Neurobiotin (0.1–0.2% in the electrode solution; Vector Laboratories, Burlingame, CA, USA) from the patch pipette. After completion of the electrophysiological recordings, the spinal cord slices were immersed overnight in 4% freshly de-polymerized formaldehyde in 0.1 M phosphate buffer (PB, pH 7.4) at 4°C, rinsed in PB, and then sectioned sagittally at 60–100 µm thickness with a Vibratome. In most cases, free-floating sections were incubated overnight at 4°C in phosphate buffered saline (PBS) with 0.3% Triton X-100 containing streptavidin-Texas Red (diluted 1: 500; Jackson ImmunoResearch, West Grove, PA, USA) and washed several times in PBS. The slices were incubated with isolectin B4 from Bandeiraea simplicifolia conjugated directly to fluorescein isothiocyanate (IB4–FITC, 0.5 µg ml–1; Sigma, St Louis, MO, USA) in order to reveal the border between laminae II and III (Silverman & Kruger, 1990; Kitchener et al. 1993; Wang et al. 1994), and then mounted in glycerol-based mounting medium (Vectashield, Vector). Sections were scanned with a LSM 510 laser scanning confocal microscope (Carl Zeiss, Oberkochen, Germany). For descriptive purposes, lamina II was divided into three equal parts: the superficial and deep thirds were defined as IIo and IIi, respectively, while the middle third was defined as the border region between IIo and IIi. This was done to allow comparison with the results of Grudt & Perl (2002). In a few cases the Neurobiotin was revealed with 3,3'-diaminobenzidine in the presence of H2O2, and sections were dehydrated, cleared and mounted in Hystomount. In these cases, the laminar location of the cell was determined by comparison with sections that had been treated with IB4.
The majority of recorded cells were located just lateral to the dorsal root entry zone, because the medial half of the SG is covered by the dorsal columns, which prevent access of the patch pipettes. Recordings from the most lateral part of the SG were avoided because the curvature of the dorsal horn in this region makes it difficult to analyse morphology of recorded cells.
Drug application
Drugs dissolved in Krebs solution were applied by exchanging solutions via a three-way stopcock without altering the perfusion rate and temperature. Drugs used were strychnine (2–4 µM; Sigma, St Louis, MO, USA), and bicuculline (20–40 µM; Sigma).
Statistical analysis
All numerical data were expressed as the mean ± S.E.M. Statistical analysis was performed on Systat 11 and StatView, and included one-way ANOVAs with post hoc tests, K-means and hierarchical cluster analysis, and discriminant analysis, as indicated in Results. Statistical significance was determined as P < 0.05. The membrane potentials were not corrected for the liquid junction potential between the Krebs and patch-pipette solutions.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Neurobiotin-labelled SG neurones resembled those described in previous studies in the rat, cat and hamster (Gobel, 1978; Bicknell & Beal, 1984; Todd & Lewis, 1986; Grudt & Perl, 2002), and could be classified into four major groups (islet, central, radial and vertical cells) according to the classification scheme proposed by Grudt & Perl (2002). The morphological classification was based mainly on dendritic arborizations (size, laminar location and major trajectory of the dendritic tree) by visual inspection and by analysis using various scatter plots of different pairwise comparisons of the parameters shown schematically (see below).
Islet cells. In our study, 20 of 133 (15.0%) cells were categorized as islet cells, because their dendritic trees were extremely elongated in the rostrocaudal direction (parallel to the layer borders) and were limited in the dorsoventral and mediolateral directions (Fig. 1A, Table 1). Cell bodies of 6 islet cells were located in lamina IIi, 2 were in lamina IIo and 12 were near the border of laminae IIi and IIo. Distinguishing between islet and central cells is often difficult (Heinke et al. 2004), although it has been reported that the rostrocaudal extent of islet cell dendritic trees is typically >400 µm (Grudt & Perl, 2002; Lu & Perl, 2003), and that axons of islet cells are generally limited to the volume of their dendritic trees (Grudt & Perl, 2002). Therefore neurones with dendritic arbors between 300 and 400 µm long were defined as islet cells if their axon remained within the volume of the dendritic tree.
|
|
Radial cells. Unlike the islet and central cells, radial cells have dendrites that extend in several directions in the parasagittal plane (i.e. they are flattened mediolaterally) (Grudt & Perl, 2002). We observed 22 (16.5%) radial cells in our sample (Fig. 1C, Table 1). The somata of 7 radial cells were located in lamina IIi, 1 was in IIo and 14 were near the border between laminae IIi and IIo.
Vertical cells. The dendritic trees of typical vertical cells emanate from somata located in lamina IIo and pass ventrally through laminae II–IV (Grudt & Perl, 2002). Therefore vertical cells have fan- or cone-shaped dendritic arbors with the soma situated dorsally at the apex. Some of these cells resemble the stalked cells described by Gobel (1975, 1978). We identified 40 (30.1%) of the neurones as vertical cells, and since the dendritic arbors of these cells had limited mediolateral spread, they showed a fan-like appearance (Fig. 1D, Table 1). Cell bodies of 27 vertical cells were located in lamina IIo, 13 were near the border between laminae IIi and IIo, and none were observed in lamina IIi.
Unclassified cells. We could not assign 27 (20.3%) SG neurones to any of the major groups described above. It has been reported previously that a relatively high proportion of neurones do not fit any of the recognized morphological categories (Todd & Lewis, 1986; Grudt & Perl, 2002; Heinke et al. 2004). Four of the unclassified neurones had rostrocaudally elongated dendritic trees similar to islet cells, but were more extensive (110–170 µm) in the mediolateral axis (Fig. 2Ca and b). Although these neurones may belong to the medial–lateral cell group (Grudt & Perl, 2002; Hantman et al. 2004), we could not confirm this even after observation in the transverse plane (Fig. 2Cc). Another morphology was observed in six unclassified neurones: these resembled central cells except that their dendritic trees arborized only on one side (rostral or caudal) of the soma. The remaining unclassified neurones showed a wide variation in shape (Fig. 2A and B).
|
|
Furthermore, in order to test whether these differences could be identified objectively, we used linear discriminant analysis through complete estimation (Prescott & De Koninck, 2002). Using four variables (RC, RC/DV, SD and SV – SD), 103 of 106 cells were correctly identified, showing that this analysis is helpful for quantitative differentiation of the four cell groups. We also examined these groupings by cluster analysis, using the K-means (Chi-square distance metric, K-means splitting cases into four groups) and hierarchical (Chi-square distance metric, Ward minimum variance method, at the level of four groups) methods. With both methods, one of the resultant clusters completely corresponded with cells in the vertical group (as determined by visual inspection), although the other three clusters were not completely matched with the remaining groups.
Subsequently, we attempted to analyse these three groups, islet, central and radial cells, by using different combinations of parameters. However we could not obtain results showing a clear identification between the three groups. We therefore excluded islet cells on the basis of their rostrocaudal dendritic extent (>400 µm) and/or the presence of an axon that was contained within the volume of the dendritic tree. We then performed a second analysis involving central and radial cells, by using two variables, SD and RC/DV (Fig. 3D). Discriminant and cluster analysis including K-means and hierarchical methods correctly identified all neurones. These data support the classification by visual inspection.
Based on the analysis of the data shown in Fig. 3, a simple and convenient scheme for distinguishing between the four major morphological classes on morphometric criteria can be summarized as follows. Cells for which RC/DV exceeds 3.5 belong to either islet or central classes. Within this group, those cells with rostrocaudal dendritic lengths >400 µm can be classified as islet cells, together with cells where this length is 300–400 µm and the axon arborizes within the volume of the dendritic tree. Cells for which RC/DV is less than 3.5 are either radial or vertical cells, and these can be distinguished on the basis of SV/SD (Fig. 3E). If this ratio exceeds 3.5 the neurone can be classified as a vertical cell, and if not, as a radial cell.
Electrophysiological experiments
Membrane properties and spontaneous excitatory and inhibitory postsynaptic currents. Table 2 provides a summary of the membrane properties of the four major cell types. There was no significant difference in resting membrane potentials among these cell types. As expected, membrane capacitance (Cm) values were highest for the islet and vertical cells, which generally had the largest dendritic trees. One-way ANOVA demonstrated a significant difference between Cm values among the four cell groups. Post hoc Bonferroni–Dunn tests showed that islet cells had significantly different values from central and radial cells, while the values for vertical cells were different from those of radial cells (P < 0.01). Similarly, membrane resistance (Rm) values were lowest for islet and vertical cells. ANOVA with post hoc Bonferroni–Dunn tests showed that values for central cells were significantly different from those of islet and vertical cells (P < 0.01).
|
|
|
- (45.5%, 5/11) and/or C- (81.8%, 9/11) fibres were observed in 90.9% (10/11) of the cells (Fig. 5A and E). However, repetitive stimulation of the dorsal roots revealed that most islet cells (72.7%, 8/11) had monosynaptic eEPSCs only from C-fibres (Fig. 5A and E). Therefore, all A-fibre-mediated EPSCs were polysynaptic. On the other hand, all islet cells (7/7) exhibited evoked IPSCs (eIPSCs) mediated by A-fibres, but not by C-fibres, when dorsal root stimulation was carried out at a holding potential of 0 mV (Fig. 5A and E). Amplitudes of eEPSCs (407 ± 67 pA) and eIPSCs (280 ± 111 pA) recorded in islet cells were significantly larger (Fig. 5F and G) than those in other three cell categories (one-way ANOVA followed by Fisher test; P < 0.05 in the post hoc test).
For central cells, dorsal root stimulation evoked C-fibre-mediated EPSCs in 72.2% (13/18) cells. In one case the EPSC appeared to be polysynaptic, while in the remaining 12 cases the eEPSCs were monosynaptic (Fig. 5B and E). As with the islet cells, we observed polysynaptic eEPSCs (16.7%, 3/18) but not monosynaptic eEPSCs from A-fibres in some central cells (Fig. 5B and E). All neurones possessed IPSCs elicited by stimulation of A- and/or C-fibres (Fig. 5B and E). One of these cells exhibited repetitive IPSCs in response to single dorsal root stimuli (see below).
In the radial cell group, monosynaptic eEPSCs from A-fibres were observed in 8 of 17 (47.1%), and polysynaptic eEPSCs from A-afferents were seen in a further 2 (11.8%) neurones (Fig. 5C and E). Eleven of 17 (64.7%) neurones had C-fibre-evoked EPSCs, and in all cases there was a monosynaptic component (Fig. 5C and E). Eight (47.1%) of the neurones received both A- and C-afferent inputs and the majority of these EPSCs were monosynaptic. The inhibitory synaptic inputs to radial neurones were similar to those of central and vertical neurones, but differed significantly from those of islet cells (Fig. 5E). A- and C-fibre-mediated IPSCs were observed in 10/13 (76.9%) and 11/13 (84.6%) radial cells, respectively (Fig. 5C and E). A remarkable characteristic of the inhibitory events observed in some SG neurones was the occurrence of repetitive IPSCs at irregular intervals that were triggered by a single dorsal root stimulus and persisted from several hundred milliseconds to a few seconds after the stimulus (Figs 5C and 6D). Another characteristic feature of this IPSC type was a relatively high proportion of failures of the initial IPSCs. This phenomenon was observed in the majority of radial cells (10/13, 76.9%; Figs 5C and 6E), and also in a small proportion of central and vertical neurones (Fig. 6E). Of the cells that showed repetitive IPSCs, 62.5% were radial cells, 12.5% were central cells and 25% were vertical cells.
|
-fibre-evoked EPSCs, and in 76.5% (13/17) of cases there was a monosynaptic component. The EPSCs in neurones that received C-fibre inputs all had a monosynaptic component (16/21, 76.2%; Fig. 5D and E). Some of these neurones received both A- and C-afferent inputs (Fig. 5D and E). As with the central and radial cells, the vertical neurones also received A-fibre-mediated IPSCs (70.8%, 17/24) and C-fibre-mediated IPSCs (70.8%, 17/24) following dorsal root stimulation (Figs 5D and E). Four of 21 vertical neurones showed repetitive IPSCs following single dorsal root stimuli (Fig. 6E).
Inhibitory transmitters of evoked IPSCs in each neurone type.
It has been reported in SG neurones that there are two types of IPSC, GABAergic and glycinergic, which can be differentiated pharmacologically and by their time courses (Yoshimura & Nishi, 1995). Since it is not possible to test all neurones with GABAA and glycine receptor amtagonists, we selected half-width as a measure of time course of eIPSCs and used this to classify the different types. In order to validate this approach, we measured the half-width of averaged eIPSCs for 18 inhibitory inputs that were tested with strychnine and/or bicuculline (glycine and GABAA receptor antagonists, respectively) (Fig. 6A–C and E). Because of the difficulty of combining this pharmacological testing with subsequent morphological analysis, some of these tests were carried out on cells for which no morphological data were available. We selected half-width as a parameter for distinguishing different types of eIPSCs because in many cases the presence of multiple superimposed IPSCs (e.g. at A- and C-fibre latencies) made it difficult to measure the decay time.
Among the cells that showed discrete (non-repetitive) IPSCs in response to dorsal root stimulation, three pharmacologically distinct groups could be identified. In one group (defined as GABA dominant), dorsal root stimulation elicited relatively long-lasting IPSCs, with a half width of 25–65 ms (n
= 10) which were blocked by the GABAA receptor antagonist, bicuculline (Fig. 6A and E). In contrast, another group of eIPSCs (glycine dominant) were short-lasting, with half-widths of 5.2–11.6 ms (n
= 5), and were blocked by the glycine receptor antagonist, strychnine (Fig. 6B and E). A third group (simple mixed) consisted of IPSCs that had both GABAergic and glycinergic components, and these had half-widths of 12.2–18 ms (n
= 3; Fig. 6C and E) (Kohno et al. 1999; Yang et al. 1999). The half-widths of all 18 averaged eIPSCs are shown in Fig. 6E, together with the cut-off values (
11.6 ms for glycinergic, >11.6 to <25.0 ms for mixed, 
25.0 ms for GABAergic). Five cells with repetitive eIPSCs were tested, and in each case both strychnine and bicuculline were needed to block the IPSCs completely. In three of these cases, strychnine was applied first and this blocked most of the currents, leaving a small residual component at fixed latency, which was bicuculline sensitive (Fig. 6D).
On the basis of these observations, we used the half-widths of eIPSCs to distinguish between the GABA-dominant, glycine-dominant and simple mixed classes, and determined the types of eIPSC that were present in cells of each morphological class (Fig. 6F). Five of the seven eIPSCs recorded from islet cells were classified as GABA dominant, one as simple mixed and the other as glycine dominant. The majority of central cells (13/14) exhibited GABA-dominant eIPSCs, with the other one showing the repetitive type. As stated above, most eIPSCs recorded in radial neurones were of the repetitive type (10 of the 11 cells that showed eIPSCs), while the remaining one was GABA dominant. Twelve of the 21 vertical cells received GABA-dominant inhibitory inputs, with the remainder being simple mixed (5) or repetitive (4).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Morphology and classification
The classification of SG neurones has proved to be a complex and controversial issue, largely because of their morphological diversity. Recent attempts to correlate morphology with transmitter content (Todd & Spike, 1993) or electrophysiological properties (Grudt & Perl, 2002) have suggested that most SG neurones can be allocated to one of four major classes: islet, central, radial and vertical cells.
One reason for the difficulty in classifying neurones is that they may show atypical features or have characteristics of more than one class. In this study, we found that although many neurones could easily be allocated to one of these classes, in other cases this was not straightforward. To avoid the need for subjective judgements, we developed a method based on the use of several morphometric parameters. This provides a more objective way of classifying neurones in those cases where this cannot be done easily on the basis of their morphological appearance.
Interpretations of neuronal networks involving each cell type based on electophysiological recordings
Figure 7 shows our interpretation of the neuronal networks involving primary afferent axons and SG neurones of each morphological type, based on our electrophysiological recordings.
|
latency. The latter were small in amplitude, and it is unlikely that a single stimulus applied to the dorsal root would cause islet cells to fire action potentials at A latency. They are therefore unlikely to contribute to A-mediated inhibition of other neurones. In all cases they received inhibitory inputs mediated by A-fibres, but not C-fibres (Fig. 7A). Islet cells are GABAergic interneurones (Todd & McKenzie, 1989; Lu & Perl, 2003; Heinke et al. 2004), and it has been suggested that the C-fibres that form synapses on them may have a relatively large diameter (Lu & Perl, 2003) and that at least some of these cells convey tactile rather than nociceptive information (Light et al. 1979; Bennett et al. 1980; Réthelyi et al. 1989). Islet cells may therefore be responsible for inhibition of dorsal horn neurones mediated by low-threshold mechanoreceptive C-fibres (Bennett et al. 1980; Lu & Perl, 2003). Islet cells do not appear to receive significant inhibitory input mediated by C-fibres, although most are excited by these afferents. It is therefore unlikely that they have extensive synaptic connections with each other, even though it has been suggested that they are linked through dendrodendritic synapses (Gobel et al. 1980; Todd, 1988). Inhibitory interneurones that receive excitatory input from A-afferents and have axons that form synapses on islet cells may be located in lamina III, since neurones with primary afferent input only from A-fibres are infrequent in SG, but are found more often in lamina III (T Yasaka, unpublished observations). A-axon terminals in laminae IIi and III are likely to belong to D-hairs, rather than nociceptors (Light & Perl, 1979), which suggests that islet cells may have powerful low-threshold mechanoreceptive inhibitory inputs.
In this study, central cells were found to have excitatory inputs mainly from C-fibres. This is generally consistent with the report by Grudt & Perl (2002) who observed monosynaptic C-fibre input in most of their central cells, while monosynaptic A input to these cells was relatively uncommon. The inhibitory inputs to central cells were mediated by both A- and C-fibres, and we found that 30% of central cells received only C-fibre-mediated inhibitory synaptic inputs (Fig. 5E). These results are compatible with the inhibitory pathway from islet cells to central cells that was reported by Lu & Perl (2003), but suggest that many central cells receive inhibitory inputs from other neurones, which are activated by A- (and possibly C-) afferents (Fig. 7B).
Radial cells generally showed both A- and C-fibre-mediated eEPSCs, a finding that is similar to that of Grudt & Perl (2002). A characteristic feature of the inhibition observed in the majority of radial cells was that glycinergic IPSCs with variable latencies occurred for up to a few seconds after a single dorsal root stimulus, while (at least in some cases) a small GABAergic IPSC was seen at a fixed latency (Figs 5C and 6D and E). One possible mechanism to explain this phenomenon is shown in Fig. 7C. Neurones that are glycinergic, but not GABAergic, are more common in deeper laminae of the dorsal horn (Todd & Sullivan, 1990), including lamina IV, which was present in the horizontal slices, and many neurones in this region respond to single sciatic nerve stimuli with a prolonged discharge of action potentials (Woolf & King, 1987). It is therefore possible that glycinergic neurones in deep dorsal horn that fire repetitively following dorsal root stimulation are responsible for glycinergic eIPSCs seen in radial cells. The GABAergic IPSCs are presumably generated by a different population of interneurones, which also receive A- and/or C-fibre input (Fig. 7C). Since the amplitudes of GABAergic eIPSCs are usually very small, it is possible that the GABAA-receptors on radial cells are distributed at extrasynaptic regions of the cell membrane.
Vertical cells showed eEPSCs mediated by A- and C-fibres (as reported by Grudt & Perl, 2002), as well as eIPSCs mediated by both classes of afferent (Fig. 7D). Most inhibitory responses recorded from vertical neurones were GABAergic, while IPSCs exhibiting both GABAergic and glycinergic components were also recorded (Figs 5D and 6A and C–F). The latter pattern of eIPSCs is shown in Fig. 6C, and was observed more often in the vertical group than in other types of SG neurone. One explanation for this type of IPSC is that GABA and glycine are released from the same presynaptic neurone, since many cells in the dorsal horn possess GABA and glycine immunoreactivities (Todd & Sullivan, 1990). Alternatively, GABA and glycine may be released from different presynaptic neurones. Further examination will be needed to distinguish between these possibilities.
In a series of pioneering studies, Perl and coworkers have begun to identify neuronal circuits involving different types of SG neurone. These include inhibitory connections from islet to central cells, and excitatory synapses from central to vertical cells and from vertical cells to lamina I projection neurones (Grudt & Perl, 2002; Lu & Perl, 2003, 2005). Our data support the idea that there is specificity in connections formed by different types of SG neurone, for example by showing that islet cells receive powerful monosynaptic input from C- (but not A-) fibres, while primary afferent-evoked IPSCs in these cells are selectively mediated by A-fibres. Our results suggest additional levels of complexity, for example by showing that other inhibitory neurones apart from islet cells are presynaptic to central cells (as well as to radial and vertical cells).
We previously reported the importance of inhibitory networks acting as a ‘lateral inhibition’ system (Kato et al. 2004). In that report, we showed that eEPSCs and eIPSCs were observed over several segments, that the rostrocaudal distribution of eIPSCs was greater than that of eEPSCs, and that action potentials recorded in SG neurones at L2 level evoked by L2 dorsal root stimulation were eliminated by inhibitory postsynaptic potentials evoked by L5 dorsal root stimulation. These results suggest that the modules existing in the dorsal horn may extend over several segments.
In conclusion, our results provide further information about excitatory and inhibitory circuits involving primary afferents and SG neurones of each morphological type. Knowledge about inhibitory circuits is necessary for understanding normal sensory processing, and is particularly important for studies of neuropathic pain, in which disinhibition is thought to play a critical role (Moore et al. 2002; Harvey et al. 2004).
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
-afferent glutamatergic transmission in substantia gelatinosa neurons of adult rat spinal cord slices. Pain 86, 273–282.[CrossRef][Medline]Baba H, Doubell TP & Woolf CJ (1999). Peripheral inflammation facilitates A fiber-mediated synaptic input to the substantia gelatinosa of the adult rat spinal cord. J Neurosci 19, 859–867.
Baba H, Ji R, Kohno T, Moore KA, Ataka T, Wakai A, Okamoto M & Woolf CJ (2003). Removal of GABAergic inhibition facilitates polysynaptic A fiber-mediated excitatory transmission to the superficial spinal dorsal horn. Mol Cell Neurosci 24, 818–830.[CrossRef][Medline]
Baba H, Yoshimura M, Nishi S & Shimoji K (1994). Synaptic responses of substantia gelatinosa neurones to dorsal column stimulation in rat spinal cord in vitro. J Physiol 478, 87–99.[Medline]
Beal JA, Nandi KN & Knight DS (1989). Characterization of long ascending tract projection neurons and non-tract neurons in the superficial dorsal horn (SDH). In Processing of Sensory Information in the Superficial Dorsal Horn of the Spinal Cord, ed. Cervero F, Bennett GJ, Headley PM, pp. 181–197. Plenum, New York.
Bennett GJ, Abdelmoumene M, Hayashi H & Dubner R (1980). Physiology and morphology of substantia gelatinosa neurons intracellularly stained with horseradish peroxidase. J Comp Neurol 194, 809–827.[CrossRef][Medline]
Bicknell HR & Beal JA (1984). Axonal and dendritic development of substantia gelatinosa neurons in the lumbosacral spinal cord of the rat. J Comp Neurol 226, 508–522.[CrossRef][Medline]
Cronin JN, Bradbury EJ & Lidierth M (2004). Laminar distribution of GABAA- and glycine-receptor mediated tonic inhibition in the dorsal horn of the rat lumbar spinal cord: effects of picrotoxin and strychnine on expression of Fos-like immunoreactivity. Pain 112, 156–163.[CrossRef][Medline]
Gobel S (1975). Neurons with axons in the substantia gelatinosa layer of the spinal trigeminal nucleus of the adult cat. Brain Res 88, 333–338.[CrossRef][Medline]
Gobel S (1978). Golgi studies of the neurons in layer II of the dorsal horn of the medulla (trigeminal nucleus caudalis). J Comp Neurol 180, 395–414.[CrossRef][Medline]
Gobel S, Falls WM, Bennett GJ, Abdelmoumene M, Hayashi H & Humphrey E (1980). An analysis of the synaptic connections of horseradish peroxidase-filled stalked cells and islet cells in the substantia gelatinosa of adult cat spinal cord. J Comp Neurol 194, 781–807.[CrossRef][Medline]
Grudt TJ & Perl ER (2002). Correlations between neuronal morphology and electrophysiological features in the rodent superficial dorsal horn. J Physiol 540, 189–207.
Hantman AW & Perl ER (2005). Molecular and genetic features of a labeled class of spinal substantia gelatinosa neurons in a transgenic mouse. J Comp Neurol 492, 90–100.[CrossRef][Medline]
Hantman AW, van den Pol AN & Perl ER (2004). Morphological and physiological features of a set of spinal substantia gelatinosa neurons defined by green fluorescent protein expression. J Neurosci 24, 836–842.
Harvey RJ, Depner UB, Wässle H, Ahmadi S, Heindl C, Reinold H, Smart TG, Harvey K, Schütz B, Abo-Salem OM, Zimmer A, Poisbeau P, Welzl H, Wolfer DP, Betz H, Zeilhofer HU & Müller U (2004). GlyR 
3: An essential target for spinal PGE2-mediated inflammatory pain sensitization. Science 304, 884–887.
Heinke B, Ruscheweyh R, Forsthuber L, Wunderbaldinger G & Sandkühler J (2004). Physiological, neurochemical and morphological properties of a subgroup of GABAergic spinal lamina II neurones identified by expression of green fluorescent protein in mice. J Physiol 560, 249–266.
Ito A, Kumamoto E, Takeda M, Shibata K, Sagai H & Yoshimura M (2000). Mechanisms for ovariectomy-induced hyperalgesia and its relief by calcitonin: participation of 5-HT1A-like receptor on C-afferent terminals in substantia gelatinosa of the rat spinal cord. J Neurosci 20, 6302–6308.
Kato G, Furue H, Katafuchi T, Yasaka T, Iwamoto Y & Yoshimura M (2004). Electrophysiological mapping of the nociceptive inputs to the substantia gelatinosa in rat horizontal spinal cord slices. J Physiol 560, 303–315.
Kitchener PD, Wilson P & Snow PJ (1993). Selective labeling of primary sensory afferent terminals in lamina II of the dorsal horn by injection of Bandeiraea simplicifolia isolectin B4 into peripheral nerves. Neuroscience 54, 545–551.[CrossRef][Medline]
Kohno T, Kumamoto E, Higashi H, Shimoji K & Yoshimura M (1999). Actions of opioids on excitatory and inhibitory transmission in substantia gelatinosa of adult rat spinal cord. J Physiol 518, 803–813.
Kumazawa T & Perl ER (1978). Excitation of marginal and substantia gelatinosa neurons in the primate spinal cord: indications of their place in dorsal horn functional organization. J Comp Neurol 177, 417–434.[CrossRef][Medline]
Light AR & Perl ER (1979). Spinal termination of functionally identified primary afferent neurons with slowly conducting myelinated fibers. J Comp Neurol 186, 133–150.[CrossRef][Medline]
Light AR, Trevino DL & Perl ER (1979). Morphological features of functionally defined neurons in the marginal zone and substantia gelatinosa of the spinal dorsal horn. J Comp Neurol 186, 151–172.[CrossRef][Medline]
Lu Y & Perl ER (2003). A specific inhibitory pathway between substantia gelatinosa neurons receiving direct C-fiber input. J Neurosci 23, 8752–8758.
Lu Y & Perl ER (2005). Modular organization of excitatory circuits between neurons of the spinal superficial dorsal horn (laminae I and II). J Neurosci 25, 3900–3907.
Moore KA, Kohno T, Karchewski LA, Scholz J, Baba H & Woolf CJ (2002). Partial peripheral nerve injury promotes a selective loss of GABAergic inhibition in the superficial dorsal horn of the spinal cord. J Neurosci 22, 6724–6731.
Nakatsuka T, Ataka T, Kumamoto E, Tamaki T & Yoshimura M (2000). Alteration in synaptic inputs through C-afferent fibers to substantia gelatinosa neurons of the rat spinal dorsal horn during postnatal development. Neuroscience 99, 549–556.[CrossRef][Medline]
Nakatsuka T, Park JS, Kumamoto E, Tamaki T & Yoshimura M (1999). Plastic changes in sensory inputs to rat substantia gelatinosa neurons following peripheral inflammation. Pain 82, 39–47.[CrossRef][Medline]
Park JS, Nakatsuka T, Nagata K, Higashi H & Yoshimura M (1999). Reorganization of the primary afferent termination in the rat spinal dorsal horn during post-natal development. Dev Brain Res 113, 29–36.[Medline]
Pearson AA (1952). Role of gelatinous substance of spinal cord in conduction of pain. Arch Neurol Psychiatry 68, 515–529.
Prescott SA & De Koninck Y (2002). Four cell types with distinctive membrane properties and morphologies in lamina I of the spinal dorsal horn of the adult rat. J Physiol 539, 817–836.
Ramon y Cajal S (1909). Histologie Du Système Nerveux de L'homme et Des Vertébrés, vol. 1. Maloine, Paris.
Réthelyi M, Light AR & Perl ER (1989). Synaptic ultrastructure of functionally and morphologically characterized neurons of the superficial spinal dorsal horn of cat. J Neurosci 9, 1846–1863.[Abstract]
Rexed B (1952). The cytoarchitectonic organization of the spinal cord in the cat. J Comp Neurol 96, 415–495.[CrossRef]
Silverman JD & Kruger L (1990). Selective neuronal glycoconjugate expression in sensory and autonomic ganglia: relation of lectin reactivity to peptide and enzyme markers. J Neurocytol 19, 789–801.[CrossRef][Medline]
Sivilotti L & Woolf CJ (1994). The contribution of GABAA and glycine receptors to central sensitization: disinhibition and touch-evoked allodynia in the spinal cord. J Neurophysiol 72, 169–179.
Todd AJ (1988). Electron microscope study of Golgi-stained cells in lamina II of the rat spinal dorsal horn. J Comp Neurol 275, 145–157.[CrossRef][Medline]
Todd AJ & Koerber HR (2005). Neuroanatomical substrates of spinal nociception. In Wall and Melzack's Textbook of Pain, 5th edn, ed. McMahon SB, Koltzenburg M, pp. 73–90. Elsevier, Edinburgh.
Todd AJ & Lewis SG (1986). The morphology of Golgi-stained neurons in lamina II of the rat spinal cord. J Anat 149, 113–119.[Medline]
Todd AJ & McKenzie J (1989). GABA-immunoreactive neurons in the dorsal horn of the rat spinal cord. Neuroscience 31, 799–806.[CrossRef][Medline]
Todd AJ & Spike RC (1993). The localization of classical transmitters and neuropeptides within neurons in laminae I–III of the mammalian spinal dorsal horn. Prog Neurobiol 41, 609–645.[CrossRef][Medline]
Todd AJ & Sullivan AC (1990). Light microscope study of the coexistence of GABA-like and glycine-like immunoreactivities in the spinal cord of the rat. J Comp Neurol 296, 496–505.[CrossRef][Medline]
Wang H, Rivero-Melian C, Robertson B & Grant G (1994). Transganglionic transport and binding of the isolectin B4 from Griffonia simplicifolia I in rat primary sensory neurons. Neuroscience 62, 539–551.[CrossRef][Medline]
Woolf CJ & King AE (1987). Physiology and morphology of multireceptive neurons with C-afferent fiber inputs in the deep dorsal horn of the rat lumbar spinal cord. J Neurophysiol 58, 460–479.
Yaksh TL (1989). Behavioral and autonomic correlates of the tactile evoked allodynia produced by spinal glycine inhibition: effects of modulatory receptor systems and excitatory amino acid antagonists. Pain 37, 111–123.[CrossRef][Medline]
Yang K, Kumamoto E, Furue H, Li YQ & Yoshimura M (1999). Action of capsaicin on dorsal root-evoked synaptic transmission to substantia gelatinosa neurons in adult rat spinal cord slices. Brain Res 830, 268–273.[CrossRef][Medline]
Yang K, Li Y, Kumamoto E, Furue H & Yoshimura M (2001). Voltage-clamp recordings of postsynaptic currents in substantia gelatinosa neurons in vitro and its applications to assess synaptic transmission. Brain Res Protoc 7, 235–240.[CrossRef][Medline]
Yoshimura M & Jessell TM (1989a). Primary afferent-evoked synaptic responses and slow potential generation in rat substantia gelatinosa neurons in vitro. J Neurophysiol 62, 96–108.
Yoshimura M & Jessell TM (1989b). Membrane properties of rat substantia gelatinosa neurons in vitro. J Neurophysiol 62, 109–118.
Yoshimura M & Nishi S (1993). Blind patch-clamp recordings from substantia gelatinosa neurons in adult rat spinal cord slices: pharmacological properties of synaptic currents. Neuroscience 53, 519–526.[CrossRef][Medline]
Yoshimura M & Nishi S (1995). Primary afferent-evoked glycine- and GABA-mediated IPSPs in substantia gelatinosa neurones in the rat spinal cord in vitro. J Physiol 482, 29–38.[Medline]
Zeilhofer HU, Studler B, Arabadzisz D, Schweizer C, Ahmadi S, Layh B, Bösl MR & Fritschy J-M (2005). Glycinergic neurons expressing enhanced green fluorescent protein in bacterial artificial chromosome transgenic mice. J Comp Neurol 482, 123–141.[CrossRef][Medline]
|
Acknowledgements |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. J. Maxwell, M. D. Belle, O. Cheunsuang, A. Stewart, and R. Morris Morphology of inhibitory and excitatory interneurons in superficial laminae of the rat dorsal horn J. Physiol., October 15, 2007; 584(2): 521 - 533. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||
